Purpose: Despite the proliferation of numerous morphometric and anthropometric methods for sex identification based on linear, angular, and regional measurements of various parts of the body, these methods are subject to error due to the observer’s knowledge and expertise. This study aimed to explore the possibility of automated sex determination using convolutional neural networks (CNNs) based on lateral cephalometric radiographs. Materials and Methods: Lateral cephalometric radiographs of 1,476 Iranian subjects (794 women and 682 men) from 18 to 49 years of age were included. Lateral cephalometric radiographs were considered as a network input and output layer including 2 classes (male and female). Eighty percent of the data was used as a training set and the rest as a test set. Hyperparameter tuning of each network was done after preprocessing and data augmentation steps. The predictive performance of different architectures (DenseNet, ResNet, and VGG) was evaluated based on their accuracy in test sets. Results: The CNN based on the DenseNet121 architecture, with an overall accuracy of 90%, had the best predictive power in sex determination. The prediction accuracy of this model was almost equal for men and women. Furthermore, with all architectures, the use of transfer learning improved predictive performance. Conclusion The results confirmed that a CNN could predict a person’s sex with high accuracy. This prediction was independent of human bias because feature extraction was done automatically. However, for more accurate sex determination on a wider scale, further studies with larger sample sizes are desirable.

Purpose: This study investigated the position of the hyoid bone and its relationship with airway dimensions in different skeletal malocclusion classes using cone-beam computed tomography (CBCT). Materials and Methods: CBCT scans of 180 participants were categorized based on the A point-nasion-B point angle into class I, class Ⅱ, and class Ⅲ malocclusions. Eight linear and 2 angular hyoid parameters (H-C3, H-EB, H-PNS, H-Me, H-X, H-Y, H-[C3-Me], C3-Me, H-S-Ba, and H-N-S) were measured. A 3-dimensional airway model was designed to measure the minimum cross-sectional area, volume, and total and upper airway length. The mean cross-sectional area, morphology, and location of the airway were also evaluated. Data were analyzed using analysis of variance and the Pearson correlation test, with p values <0.05 indicating statistical significance. Results: The mean airway volume differed significantly among the malocclusion classes (p<0.05). The smallest and largest volumes were noted in class Ⅱ (2107.8±844.7 mm3) and class Ⅲ (2826.6±2505.3 mm3), respectively. The means of most hyoid parameters (C3-Me, C3-H, H-Eb, H-Me, H-S-Ba, H-N-S, and H-PNS) differed significantly among the malocclusion classes. In all classes, H-Eb was correlated with the minimum cross-sectional area and airway morphology, and H-PNS was correlated with total airway length. A significant correlation was also noted between H-Y and total airway length in class Ⅱ and Ⅲ malocclusions and between H-Y and upper airway length in class I malocclusions. Conclusion The position of the hyoid bone was associated with airway dimensions and should be considered during orthognathic surgery due to the risk of airway obstruction.